ML20059C112
| ML20059C112 | |
| Person / Time | |
|---|---|
| Issue date: | 10/31/1993 |
| From: | NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | |
| References | |
| NUREG-BR-0112, NUREG-BR-0112-V06-N2, NUREG-BR-112, NUREG-BR-112-V6-N2, NUDOCS 9311010018 | |
| Download: ML20059C112 (8) | |
Text
- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _
NUREGIBR-0112 l#
@**$RESEARCH NEWS OFFICE OF NUCLEAR REGULATORY RESEARCH U.S. NUCLEAR REGULATORY COMMISSION VOLUME 6, NUMBER 2 OCTOBER 1993 Operability Research on motor.
Technical Efforts Related to Resolving GSI-87 Operated Valves initial research sponsored by DE/RES and com-pleted.in 1987 revealed that adequate motor-Gerald H. Weidenhamer, DE/EMEB operated valve (MOV) qualification testing at high-
- flow, high-pressure, and high-temperature conditions had not been performed. The MOV Regulatory issue testing completed in the early 1980s by the indus-try was mainly go/no-go type testing on small
]
in 1986, the Division of Engineering (DE), RES, valves. Based on the results of tho limited industry 4
started work to address a safety issue that had testing, DE/RES concluded that additional tests been identified by the Advisory Committee for would be necessary to determine whether these Reactor Safeguards (ACRS). The ACRS was con-MOVs will close as they should under high-flow I
cerned that a full break in certain high-energy (blowdown) conditions. Over the next 2 years pipes outside the containments of boiling water (1988-89), DE/RES sponsored MOV flow tests to reactor (BWR) plants, coupled with failure of the develop the technical basis for assessing MOV ca-isolation valves in these pipes to close, would re-pabilities. These tests were conducted on compo-sult in unacceptable consequences. The flow from nents typical of those installed in these pipes, It one of these broken pipes,if unchecked, can have was intended that the results of these tests would serious effects not only because of a steam re-provide the technical basis for resolving GSI-87.
lease outside containment, but also because oth-er emergency equipment (pumps, electrical com-A total of six different gate valves were tested-ponents, etc.) may be exposed to harsh water and three with 6-inch diameters and three with 10-inch l
steam environments and fail. These components diameters. The 6-inch valves are typical of those outside containment are not required to be quali-installed in the RWCU pipes, and the 10 inch 7
fied to these harsh environments. This ACRS con-valves are typical of those installed in the HPCI cern was identified as a high priority Generic Safe-steam pipes. Some MOVs in the RCIC cystem are ty issue, GSI-87, " Failure of the High Pressure typical of those in the RWCU system except that Coolant injection (HPCI) Steam Line Without iso-fluid environments are different. All fluid environ-lation." The concern was expanded to include the ments (temperature, pressure, flow velocity) for reactor core isolation cooling (RCIC) steam lines hardware in the three systems were simulated to as well as the reactor water cleanup (RWCU) hot-conform to actual conditions that would occur in a water lines in BWRs. These pipes all penetrate the real pipe break accident situation for those GSI-87 l
primary containment wall, and typically have nor-systems. The sketch shows the main parts of a i
l mally open isolation valves installed in the pipes-typical motor-operated gate valve similar to those
)
l one inside the containment building and the other installed as isolation valves that are discussed in
(
outside the containment building.
this article.
1 9311010018 931031 PDR NUREG BR-0112 R PDR.
l
_ _ _ _ _ ___ _ _ ___ __ ___ _____ _ _ U
~
P i
Research Results F
{
The main findings from the blow-down tests were
~
Electric motor- --
.==
ing the highest pressure blowdown test because the selected torque switch position had been un-t
/
4 derestimated for this high load condition. Full clo-4 sure was reached on all previous and subsequent a
- 9 tests of this valve. One of the other 6-inch valves yg reached full closure during the blowdown test, but -
I ii-the thrust measurement was different from that ex-pected. A subsequent inspection of the valve inter-nals revealed excessive damage to the guide I
( l, channels on the disc and to the disc guides 10-
=
==c cated on the valve body. The disc was refurbished
'Nr and retested under the same conditions at a later valve stem -
seat rino
,.77;fq date; however, this time the valve did not reach full closureandhadexcessiveleakage. Asubsequent Disc ' 'Q inspection revealed that damage had occurred on -
L the downstream sealing surfaces of the disc and the housing. It was concluded that the failure to ful-ly close and the damage to the sealing surfaces Typical Motor-Operated Gate Valve both contributed to the leakage problem. Because of the damage, this valve was not used in any fur-ther testing. The other four valves were capable of stopping the high-velocity flows in all the closing tests. One of the 10-inch valves also experienced -
One of the main goals of the RES-sponsored tests significant damage during closure, but the dam-was to obtain sufficient data to permit the quantifi-age did not prevent flow isolation for this valve. It is cation of all force terms contained in the accepted important to recall that all valve motor operators 1
industry equation used for predicting the thrusts had been set to deliver larger thrusts than would i
required to operate these MOVs. Therefore, test normally have been the case for valves in operat-procedures were developed, and specific param-ing plants. Another important finding from the tests eters were identified for measurements for a quan-is that the internal MOV friction forces as calcu-tification capability. These procedures included lated from the industry formula are underpre-setting the torque switches of the motor-operators dicted. This latter finding leads to either undersiz-to deliver sufficient thrusts to ensure closure of the ing the motor-operators or incorrectly setting the ~
valves without overstressing either the operator or torque switches of the motor-operators for these the MOV components. Employing these proce-serious events. These results indicated to the N RC
=l dures and recording the speciC0 parameters al-that some MOVs in operating plants probably lows all forces, including the internal friction forces would not have been capable of isolating flows if a -
of the valves, to be calculated over the entire oper-full guillotine break occurred in one of the specific j
ating strokes of the valves. The resulting thrust cal-GSI-87 pipes. The NRC regulatory staff took action ~
culations could then be directly compared with the by requesting information on these MOVs from the measured thrusts, licensees. The licensees were notified where spe-cific MOVs were' either believed to be under-powered or where torque switches were set too These procedures and the subsequent data inter-low. Ucensee actions have been taken to address pretations have provided it.e technical break-these NRC concerns.
through for understanding the complicated behavior of specific types of MOVs under large-During the 2 years following the tests, most of the flow-load accident conditions.
MOV test data were analyzed, leading to the 2
i development of an improved method for estimat-plants are covered by GL 89-10. The licensees'-
ing the required closing thrust for the GSI-87 gate valve engineers are testing some MOVs and are valves and for other gate valves subjected to vari-using the results for setting the torque switches of ous flow-loading conditions. With these test data, their MOVs to comply with GL 89-10. The Electric it was also possible to quantify the friction factors Power Research Institute (EPRI) is also conduct-internal to the valves that are dependent on fluid ing a large MOV research program to develop ana-properties and important for accurately estimating lytical methods to assess MOV operability. When the required closure forces. A user-friendly com-completed, the EPRI results will be used by mem-puter program was also developed as part of this ber utilities to set up MOVs in their plants to comply effort and is an aid in simplifying these compil-with GL 89-10. The NRC is continuing to review cated calculations.
the EPRI research efforts and will evaluate the re-ports of the results as they become available.
The results from the flow tests provided the basis for resolving GSI-87 in 1992.
NRC research results are primarily for the use of NRC regulators and inspectors, for assessing indi-vidual plant MOV programs and for evaluating the Use of Research Results for Other Regulatory capability of specifidn-plant MOVs. However, the Applications results are also being utilized by many licensees
~
for setting up similar MOVs. In addition, these re-suits have provided the technical basis for the is-As a result of continued poor MOV performance suance of several regulatory documents, includ-from 1985 to 1989, and because of the findings ing supplements to GL 89-10 and information from the RES-sponsored MOV tests discussed Notices.
previously, the NRC issued Generic Letter GL 89-10, " Safety-Related Motor-Operated Valve Testing and Surveillance," in June 1989. This doc-Research on MOVs is continuing, and the results f
ument requests that licensees develop an overall are expected to advance the state of the art in MOV program that would lead to ensuring MOV capabil_
technology and provide a basis for evaluating li-ity. The generic letter states that licensees are to censees' implementation of GL 89-10. One area identify the design basis challenge that each MOV being investigated at this time is the effects of ag-covered by GL 89-10 is to operate against, in ad.
ing on thrust requirements after a valve has been in dition, licensees are encouraged to determine the a corrosive environment over a period of time.
torque-switch set points bytesting the MOVs atthe This investigation includes small-sample corro-design basis conditions, where practicable. Other sion tests on typical valve materials and will be recommendations are also included in GL 89-10.
completed in 1993. Other areas being researched include developing a method for prioritizing MOVs Although the MOVs tested and discussed pre-based on risk importance, developing a basis for viously were for specific GSI-87 piping systems determining motor-operator margins, and investi-and fluid conditions (high flows), these same gating degraded voltage concerns.
MOVs also had been tested at lower flow, lower temperature, and lower pressure conditions while in all cases, the results of the MOV tests and ad-each MOV was installed in the test stands. This is vances in knowledge are being discussed with in-particularly important because it permits the re-dustry MOV engineers and consensus standards suits and knowledge gained from these latter tests committees on MOV qualification and in-service to be applicable to a large number of MOVs in testing. Two standards are currently being revised other systems within the plants. Therefore, these to reflect some of these findings. The RES MOV re-overall results are applicable to many of the search, the industry's efforts, and the EPRI MOV safety-related MOVs identified by GL 89-10.
research efforts are all helping to provide technical bases for understanding MOV behavior and are it has been estimated that nearly 15,000 MOVs (in-contributing to improving MOV reliability and over-ciuding the GSI-87 MOVs) in currently operating all nuclear plant safety.
l 3
7 v-,-
-w,y v
7---
,c,
. ny -, -,- r
Acknowledgment it is important to acknowledge the work and efforts of Robert Steele, Kevin DeWall, and John Watkins The XR1-1 ex-reactor experiment was successful-of the Idaho National Engineering Laboratory, ly performed at Sandia National Laboratories on each of whom has made signif; cant contributions August 1,1993. XR1-1 was the first of a series of to the work reported here. The RES Division of ex-reactor experiments to determine, for the Safety issue Resolution also supported parts of range of BWR dry core accident conditions pro-this program during the early stages.
duced by primary system depressurization, whether metallic melt (BWR control blade material and zircaloy from the core) drains from the reactor core or freezes to form a core blockage (as at Tw mty-First Water Reactor Safety TMI-2). Metallic melt drainage from the core and information Meetiri9 BWR core plate leads to a low-temperature metal-lic melt on the vessel lower head, while melt-The Twenty-First Water Reactor Safety information through from a blocked core, as at TMi-2, pro-Meeting will be held on October 25-27,1993, 8:30 duces a large pour of hot ceramic (and fuel) melt a.m. to 5:00 p.m., in the Bethesda Marriott Hotel, onto the lower head. These two different severe 5151 Pooks Hill Road, Bethesda, Maryland.
accident pathways significantly affect the proba-bility and the mode of reactor vessel failure dunng a core melt accident.
The purpose of this annual meeting is to review progress and technical accomplishments in re-search programs. The meeting includes papers The test section for these experiments consists of a full-scale radial and axial section of the lower and discussions covering the status of research programs. The intemational meeting includes par-quarter of the central region of a BWR core and ticipation by personnel from U.S. Government lab-core plate with a prototypic axial temperature dis- -
oratories, various research firms and independent tribution. In these experiments, metallic melt of laboratories, reactor vendors, util prototypic composition and temperature that rep-ties, and a number of foreign countnl ties, universi-resents the metallic melt draining from the hot es. This meet-ing is sponsored by the NRC and conducted by upper three-fourths of the core is poured at a pro-the Brookhaven National Laboratory.
totypic rate (a dribble) into the top of the 1-meter-long test section. The primary purpose of the two initial XR1 experiments (see Figure 1) is to learn The preliminary agenda for the Water Reactor how to perform the very difficult XR experiments Safety information Meeting includes 17 sessions and to work out the bugs before going on the the.
on Severe Accident Research, Primary System In-more complex and expensive XR2 experiments.
tegrity, Advanced Reactor Research, Thermal i
Hydraulics, Aging Research, Products and Appli-cations, Advanced Control System Technology, After conduct of the XR1-2 test (steeper axial ther-Advanced instrumentation and Control Hardware, mal gradient), the test sections will be modified to and Human Factors Research.
include fuel rods within the channel box regions, as shown in Figure 2. These tests, designated XR2, will include significantly more complete geo-Attendees may register at the medng or may reg-metrical representations of the lower BWR core re-ister in advance by contacting Susan Monteleone, gion, including core plate, nosepieces, and fuel
)
Brookhaven National Laboratory, Department of support pieces, and will involve a two-stage melt Nuclear Energy, Building 130, Upton, NY 11973, delivery of, first, molten control blade materials, i
Telephone (516) 282-7235; or Christine Bonsby, followed by molten zircaloy core materials. It 1
Office of Nuclear Regulatory Research, U.S. Nu-should be emphasized that although useful phe-clear Regulatory Commission, Washington, DC nomenological test results are anticipated from 20555. Telephone (301) 492-3618.
the simple-channel XR1 tests, the intent of these 4
XR1-1 Simple Channel Geometry The development of the system for the slow, con-gg* a trolled delivery of the hot metallic melt has been a I$8IMN.e$b;%troiks major problem in the development of the XR ex-
.m p Wppa$gggyg[g,$
periments and has caused a schedule slippage of
-g$
y about 6 months. Performance of the initial XR1-1 M /
experiment, which included the first complete test d
%. th.5;@ M.
of the new melt delivery system, was completed
- i. -
'"" 3Mn~rAlwhis'$
g g w r u e n $-
successfully. Allsystemsperformedsatisfactorily, g
including the on-line X-ray monitor of the melt be-havior. Little or none of the control blade melt
.e penetrated to the bottom of the 1 meter test sec-tion.The posttest sectioning of the XR1-1 test sec-tion should provide good data on the relevant ma-terials interactions and on the melt behavior in the g
boron carbide control-blade / channel-box system.
ZrO, insulation Maintenance Rule Revision Figure 1 oss-Se i nalViev of the On June 23,1993, a final rule was published in the Federa/ Register (58 FR 33993) amending 10 CFR 50.65, " Monitoring the Effectiveness of Mainte-initial experiments was to develop the technology nance at Nuclear Pownr Plants." The final rule mo-required for conducting the more complicated difies the period for the performance of evalua-XR2 experiments.
tions required by power plant licensees from annually to once per refueling cycle, but not to ex-ceed 2 years. Because of the quality and amount i
XR2 Channel Geometry with Fuel Rods of data required, this change will provide greater i
assurance for ' effective maintenance and im-
__g ZrO, insulation control blade proved plant safety.
zr
'h*j 7 NRC Joins Large-Scale Seismic E ----
,,, W \\ cx. -
Test Program in Taiwan m u m uuuuo y
00L f66bbdbbOF'*
H. L. Graves, DE/SSEB 00 lO000000 3
- OO
- 000000 On April 24,1993, the Large-Scale Seismic Test j
- OO l
' OOOOO (LSST) facility was dedicated in a ribbon-cutting j
- OO OOOO sn ny in Hualien, T iw n Republic of China.
- OGO Senior managers frorn the eleven organizations
, OO i
OO ;'
participating in this cooperative effort attended the
- O ceremony. The dedication ceremony was covered 10 0
- 0 by a press corps representing local newspapers i
tw.
and nuclear industry publications from Taiwan and fuel Japan.
l rods The LSST Program was initiated in January 1990, and is expected to continue for 5 years. The goal of Figure 2 Cross-Sectional View of the DR2 this program is to collect real earthquake-induced Test Channel with Fuel Rods soil structure interaction (SSI) data in order to 5
evaluate computer codes used in SSI analysis of nuclear power plant structures. In the program, j#,,, A' observations will be made on the motions of the i
s reactor building model and the surrounding g~t-[-g*- -} -
ground during large-scale earthquakes. The ex-g
'-'I pectation is that the test model will be shaken by I
\\
numerous earthquakes in this seismically active
- - - m}-]'
[ jlf h area of Taiwan. Instrumentation located on the y
scale model and in the field along a three-j dimensional strong ground motion array, see Fig-l ure 1, will record any observed data. The LSST Program at Hualien, Taiwan, is a follow-on to the a
Soil-Structure Interaction (SSI) experiments at Lo-l tung, Taiwan.
~
Figure 2 Plan and Section of Test Model
.uu ouu:
The Electric Power Research Institute (EPRI) and
\\
- 7,'~,,'"*,*"~~~,',
the Taiwan Power Company (Taipower) organized the LSST program and took the lead in planning e-and managing the program. Other organizations
'[q that provided cofunding and technical as well as a-
=
l y
i1 management input to the LSST Program are the
' ~ ' ~ *
'E' crq NRC, the Central Research Institute of Electric J
Power Industry, the Tokyo Electric Power Compa-i P j ;s 1
7*" "
ny, the Commissariat a l'Energie Atomique, the Electricite de France, Framatome, the Korea Elec-f M's s%
tric Power Corporation, the Korea Institute of Nu-i clear Safety, and the Korea Power Engineering 1
Company.
A 17-minute VHS video tape titled "The Forced
[_._*_ l Vibration Test for Large-Scale Building Model in Hualien, Taiwan," is available for viewing; to view Figure 1 Free-Field Instrumentation Layout the tape please phone Herman Graves at (301) for the Hualien LSST Project.
492-3860. The video highlights LSST progress through April 1993 and shows the forced vibration tests being conducted.
The LSST facility is one of the largest in the world Release of MELCOR 1.8.2 for SSI research. The construction of a 1/4-scale model of a reinforced concrete containment,10.5 A period of extensive modification and updating of the NRC's severe accident analysis code MEL-meters in diameter and 16.5 meters high (11.1 me-COR has been completed with the release of ters above the ground), has been completed, see MELCOR version 1.8.2. The MELCOR code mod-Figure 2.
els the entire progression of severe accidents in 6
light-water cooled nuclear power plants, including that the peak pressure increase measured in the the reactor coolant system and containment ther-Surtsey vesselwas 0.2 MPa (29 psia). During the mal and hydraulic response, core heatup and deg-test, the video camera showed very little debris radation, fuel and core relocation, and fission entrainment into the upper containment atmo-product release and transport for both BWRs and sphere. Posttest sample (de'oris, hydrogen, etc.)
PWRs.
analysis is ongoing.
The recent improvements include modeling of di-rect containment heating following high-pressure SBWR Test Facility melt ejection, modeling of containment ice con-densers, modeling of reactor vessellower plenum Purdue University was selected as the contractor debris behavior and bottom head heatup and fail-to design, build, and operate an NRC-sponsored ure, modeling of radial spreading of molten and integral simplified boiling-water reactor (SBWR) particulate debris, modeling high-temperature eu-test facility. The integral SBWR test facility at Pur-tectic reactions and dissolution of solids in molten due will be a low-pressure, quarter-height scaled mixtures, and many changes to correct operation-model of the SBWR with a volume scale of 1/400. It al difficulties of varying degrees of severity.
will have all the key components and systems re-quired for investigating the integral performance of MELCOR 1.8.2 was reviewed by a peer group, an the gravity-driven cooling system and the passive expert review panel of eight people, who per-containment cooling system, which are the two formed a thorough evaluation of the adequacy of passive safety systems unique to the SBWR. Test the code and its documentation.
results from this integral facility will be used not only to confirm GE's data from other test facilities, but also to broaden the data base for assessing Direct Containment Heating Experi-SBWR codes for accident analyses.
mentS The final integral effects test, using a 1/10-scaled RESEARCH NEWS is published by the USNRC Of-representation of the Surry Nuclear Generating fice of Nuclear Regulatory Research, Ann F.
Beranek, Editor.
Station at the Sandia National Laboratory Surtsey facility, was successfully conducted recently.
Comments, suggestions, and articles for future is-sues should be directed to the Editor, RESEARCH Iron oxide / aluminum thermite was used as a co-NEWS, Office of Nuclear Regulatory Research, rium melt simulant. Thirty kilograms of molten U.S. Nuclear Regulatory Commission, Washing-thermite were ejected by superheated steam at a ton, DC 20555.
pressure of 12 MPa through the hole in the melt generator into the scale model of the Surry-like Copies of NRC and other Govemment publica-cavity. The cavity and the containment basement tions may be purchased from the Govemment floor contained no water. Just before high-Printing Office at the current GPO price. Informa-pressure melt ejection, the containment vessel tion on current GPO prices may be obtained by was pressurized to 0.16 MPa (24 psia) at a temper-contacting the Superintendent of Documents, ature of ~400K (127'C) using a mixture com-U.S. Government Printing Office, Post Office Box posed of air (43 mol.%), steam (52 mol.%), and 37082, Washington, DC 20013-7082, telephone hydrogen (5 mol.%). The preliminary data showed (202) 512-2249 or (202) 512-2171.
7
UNITED STATES NUCLEAR REGULATORY COMMISSION po3[gGE AND ST E S PAID WASHINGTON, D.C. 20555-0001 uSnac PERMIT NO. G 67 OFFICIAL BUSINESS PENALTY FOR PRIVATE USE. $300 12055513'3531.
1 1A01A91"L1 R hy of PUBLICATIONS SVCS TPS-POR-NUREG P-211 nC 23555 yasHINGTON
-